Signal optimization result evaluation method based on multiple indexes

Abstract

The invention provides a signal optimization result evaluation method based on multiple indexes, and the method comprises the steps: obtaining an optimization target of the last traffic intersection signal lamp timing from a historical record in response to a target signal; according to the optimization target of the last traffic intersection signal lamp timing, determining a plurality of evaluation indexes corresponding to the optimization target of the traffic intersection signal lamp timing; parameters corresponding to the evaluation indexes are obtained based on the multiple evaluation indexes, and the parameters are obtained based on traffic flow and / or people flow data of the traffic intersection; and according to the plurality of evaluation indexes and the corresponding parameters, evaluating the signal lamp timing optimization result of the traffic intersection to obtain the evaluation result. By implementing the method and the device, the verification automation is improved, the optimization direction of signal lamp timing can be found or adjusted in time, the timing of the signal lamp can be readjusted, the passing efficiency is improved, the evaluation is more targeted, and the evaluation is more comprehensive and the evaluation result is more accurate due to multiple evaluation indexes.

Description

Technical Field

[0001] The invention belongs to the technical field of data processing, and in particular relates to a signal optimization result evaluation method based on multiple indicators. Background Art

[0002] In modern urban traffic management, traffic light timing optimization is a key means to improve road use efficiency, reduce traffic congestion, and enhance driving and pedestrian safety. With the rapid development of urbanization, traffic flow is increasing. In order to ease traffic pressure, a traffic light timing optimization method is proposed to optimize the timing of traffic lights to adapt to dynamically changing traffic needs. In related technologies, it is impossible to verify whether the traffic light timing optimization method meets the current traffic conditions, resulting in the traffic light timing optimization method not being applicable to the current traffic conditions, causing unreasonable traffic light timing and reducing traffic efficiency. Summary of the invention

[0003] In view of this, the object of the present invention is to provide a signal optimization result evaluation method based on multiple indicators to meet the needs of verifying whether the signal light timing optimization method meets the current traffic conditions.

[0004] In order to achieve the above object, the present invention provides the following technical solutions:

[0005] According to the first aspect, the present invention provides a signal optimization result evaluation method based on multiple indicators, comprising: in response to a target signal, obtaining the optimization target of the last traffic intersection signal light timing from historical records; determining multiple corresponding evaluation indicators based on the optimization target of the last traffic intersection signal light timing; based on the multiple evaluation indicators, obtaining parameters corresponding to the evaluation indicators, the parameters are obtained based on vehicle flow and / or pedestrian flow data at the traffic intersection; based on the multiple evaluation indicators and their corresponding parameters, evaluating the signal light timing optimization result at the traffic intersection to obtain this evaluation result.

[0006] Optionally, the method of determining multiple corresponding evaluation indicators based on the optimization target of the last traffic intersection signal light timing includes: performing natural language processing on the optimization target to obtain at least one keyword; when the number of keywords is less than the target number, expanding the keywords to obtain multiple related recommendation words; matching multiple first evaluation indicators corresponding to the multiple keywords and / or related recommendation words; inputting parameters corresponding to the multiple first evaluation indicators into the digital twin model to simulate the traffic intersection situation; adjusting the traffic intersection signal light timing to obtain the correlation between the first evaluation indicators; merging multiple first evaluation indicators whose correlation exceeds a first preset threshold to obtain multiple evaluation indicators corresponding to the traffic intersection signal light timing optimization target.

[0007] Optionally, after calculating the correlation of the multiple first evaluation indicators, it includes: marking the multiple first evaluation indicators whose correlation calculation is lower than the second preset threshold value to obtain the target first evaluation indicator, and evaluating the optimization result of the signal light timing at the traffic intersection according to the multiple evaluation indicators and their corresponding parameters to obtain the current evaluation result, including: when there are multiple target first evaluation indicators, establishing a strategy variable combination of multiple target first evaluation indicators according to the particle swarm optimization algorithm; constructing a target space according to the strategy variable combination, the target space contains all strategy variable combinations; identifying the Pareto front in the target space; determining the best decision result in the Pareto front according to preset constraints; quantifying the difference between the parameters corresponding to the target first evaluation indicator of the best decision result and the parameters corresponding to the target first evaluation indicator in the optimization target to obtain a differentiated result; and determining the final result of the optimization target of the signal light timing at the last traffic intersection in this evaluation according to the differentiated result as the current evaluation result.

[0008] Optionally, the difference quantification of the target first evaluation indicator corresponding parameters of the best decision result and the target first evaluation indicator corresponding parameters in the optimization target to obtain a differentiated result includes: taking the target first evaluation indicator corresponding parameters of the best decision result as the first sequence, and taking the target first evaluation indicator corresponding parameters in the optimization target as the second sequence; performing dimensionless processing on the data in the first sequence and the data in the second sequence; determining the difference value between each data in the second sequence after the dimensionless processing and the data in the first sequence according to the maximum difference and the minimum difference between the data in the first sequence after the dimensionless processing and the data in the second sequence; and obtaining a differentiated result according to the difference value and its corresponding weight.

[0009] Optionally, when there are multiple unmarked first evaluation indicators, the final result of the optimization target of the signal timing of the previous traffic intersection in this evaluation is determined based on the differentiated result as the result of this evaluation, including: determining the positions of the corresponding parameters of the multiple unmarked first evaluation indicators in the evaluation indicator parameter interval preset for the optimization target; determining the evaluation values ​​of the multiple unmarked first evaluation indicators based on the positions of the corresponding parameters of the multiple unmarked first evaluation indicators in the interval; obtaining the evaluation results of the multiple unmarked first evaluation indicators based on the preset first evaluation indicator weights and the corresponding evaluation values; obtaining the differentiated evaluation results based on the differentiated result and its corresponding first evaluation indicator weights; obtaining the final result of the optimization target of the signal timing of the previous traffic intersection in this evaluation as the result of this evaluation based on the evaluation results of the multiple unmarked first evaluation indicators and the differentiated evaluation results.

[0010] Optionally, a signal optimization result evaluation method based on multiple indicators also includes: obtaining multiple evaluation results; performing trend analysis on the multiple evaluation results to obtain a trend analysis graph, wherein the trend analysis graph includes points corresponding to the multiple evaluation results and an overall trend; and determining, based on the trend analysis graph, the applicability of the optimization target of the signal timing at the last traffic intersection under the current traffic intersection conditions.

[0011] Optionally, the trend analysis graph includes points corresponding to multiple evaluation results and overall trend data, and determining the applicability of the optimization target of the signal light timing at the previous traffic intersection under the current traffic intersection conditions based on the trend analysis results includes: obtaining points corresponding to multiple evaluation results of the trend analysis graph, and determining slope data of the line between the point in the trend analysis graph and adjacent points; determining a first suitability parameter based on the slope data of the line between the point in the trend analysis graph and adjacent points; determining a second suitability parameter based on the overall trend data; and determining the applicability of the optimization target of the signal light timing at the previous traffic intersection under the current traffic intersection conditions based on the first suitability parameter and the second suitability parameter.

[0012] According to the second aspect, an embodiment of the present invention provides a signal optimization result evaluation device based on multiple indicators, including: a target acquisition module, used to obtain the optimization target of the last traffic intersection signal light timing from historical records in response to a target signal; an indicator determination module, used to determine the corresponding multiple evaluation indicators based on the optimization target of the last traffic intersection signal light timing; a parameter acquisition module, used to obtain parameters corresponding to the evaluation indicators based on the multiple evaluation indicators, and the parameters are obtained based on the vehicle flow and / or pedestrian flow data at the traffic intersection; a result determination module, used to evaluate the signal light timing optimization result at the traffic intersection based on the multiple evaluation indicators and their corresponding parameters, and obtain this evaluation result.

[0013] According to the third aspect, an embodiment of the present invention provides an electronic device, comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the steps of the signal optimization result evaluation method based on multiple indicators as described in the first aspect or any implementation scheme of the first aspect.

[0014] According to the fourth aspect, an embodiment of the present invention provides a computer storage medium having computer instructions stored thereon, which, when executed by a processor, implement the steps of the signal optimization result evaluation method based on multiple indicators as described in the first aspect or any embodiment of the first aspect.

[0015] The embodiment of the present invention provides a signal optimization result evaluation method based on multiple indicators. When a target signal is received, it automatically starts to verify whether the signal light timing optimization method meets the current traffic conditions, thereby improving the automation of verification, facilitating timely discovery or adjustment of the optimization direction of the signal light timing, and readjusting the signal light timing to improve traffic efficiency. In addition, the embodiment of the present invention determines the corresponding evaluation indicators based on the optimization target of the signal light timing at the last traffic intersection, making the evaluation more targeted, and multiple evaluation indicators make the evaluation more comprehensive and the evaluation results more accurate.

[0016] Other advantages, objectives and features of the present invention will be described in the following description and will be apparent to those skilled in the art to some extent, or those skilled in the art may be taught from the practice of the present invention. The objectives and other advantages of the present invention may be realized and obtained through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to make the purpose, technical solution and beneficial effects of the present invention clearer, the present invention provides the following drawings for illustration:

[0018] Figure 1 A specific example flow chart of a signal optimization result evaluation method based on multiple indicators of the present invention;

[0019] Figure 2 It is a principle block diagram of a specific example of an electronic device in an embodiment of the present invention. DETAILED DESCRIPTION

[0020] The technical solution of the present invention will be described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0021] In the description of the present invention, it should be noted that, unless otherwise clearly specified and limited, the terms "installed", "connected", and "connected" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, it can also be the internal connection of two components, it can be a wireless connection, or it can be a wired connection. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0022] In addition, the technical features involved in the different embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0023] This embodiment provides a signal optimization result evaluation method based on multiple indicators, such as Figure 1 As shown, including:

[0024] S101, in response to a target signal, obtaining an optimization target for the last traffic intersection signal timing from historical records;

[0025] S102, determining a plurality of corresponding evaluation indicators according to the optimization target of the last traffic intersection signal light timing;

[0026] S103, based on the multiple evaluation indicators, obtaining parameters corresponding to the evaluation indicators, where the parameters are obtained based on vehicle flow and / or pedestrian flow data at the traffic intersection;

[0027] S104, evaluating the signal light timing optimization result at the traffic intersection according to the multiple evaluation indicators and their corresponding parameters to obtain the current evaluation result.

[0028] For example, the target signal can be when the sensors and monitoring equipment deployed at the traffic intersection detect a traffic accident or severe congestion. Severe congestion can be characterized by the duration of congestion, for example, when the duration of continuous congestion exceeds 20 minutes, that is, when a traffic accident or severe congestion occurs at the traffic intersection, the evaluation algorithm is started to complete the evaluation of the optimization results of the signal light timing. The target signal can also be when the timer reaches the timing time, for example, the timing is one week. When the timer reaches one week, the target signal is issued to start the evaluation algorithm and complete the evaluation of the optimization results of the signal light timing.

[0029] The optimization target of the last traffic intersection signal timing may be one or more combinations of 100% driving safety, 70% traffic capacity, 100% pedestrian safety, public transportation priority, etc. This embodiment does not limit the optimization target, and those skilled in the art can determine it according to actual conditions.

[0030] According to the optimization target of the last traffic light timing at a traffic intersection, determining multiple evaluation indicators corresponding to the optimization target of the traffic light timing at the traffic intersection can be by retrieving a mapping relationship table of optimization targets and evaluation indicators pre-stored in the system. Obtain multiple evaluation indicators according to the mapping relationship table. The method of obtaining parameters corresponding to the evaluation indicators can be to obtain traffic and / or pedestrian flow data at the traffic intersection through sensors or monitoring equipment, and then clean the obtained data to obtain parameters corresponding to the evaluation indicators. The parameters corresponding to the evaluation indicators represent the parameters used to calculate the results corresponding to the evaluation indicators. For example, when the evaluation indicator is traffic capacity, its parameters are the number of vehicles passing through the traffic intersection within a specific time.

[0031] After obtaining multiple evaluation indicators and corresponding parameters, the evaluation method of the evaluation indicator is determined. In this embodiment, the difference between the parameter corresponding to the evaluation indicator and the evaluation indicator parameter preset by the optimization target can be used, and normalization processing is performed according to the difference to obtain the evaluation result corresponding to the evaluation indicator. For example, for the optimization target of reaching 70% of the traffic capacity, its parameter is the number of vehicles passing through the traffic intersection within a specific time, that is, a quantifiable parameter. Taking the number of vehicles passing through the traffic intersection within a specific time as 380 as an example, the evaluation result of this indicator of traffic capacity is calculated in the following way: first, the maximum traffic capacity of the traffic intersection is determined. For example, the maximum number of vehicles passing through the traffic intersection within a specific time is 800. Then, to achieve the optimization target, at least 560 vehicles need to pass through. For the number of vehicles passing through the traffic intersection within a specific time is 380, the evaluation result of the traffic capacity indicator is calculated by dividing the number of vehicles actually passing through the traffic intersection within a specific time by the number of vehicles that need to pass through the traffic intersection at least to achieve the optimization target, that is, 0.68. After obtaining the evaluation results of all evaluation indicators according to the above method, the evaluation result is obtained by weighted summing each evaluation indicator result and the corresponding weight. This embodiment is described by taking the above evaluation result as an example, and it can also be implemented by using technologies such as neural networks. This embodiment does not limit the specific method of obtaining the evaluation result.

[0032] The embodiment of the present invention provides a signal optimization result evaluation method based on multiple indicators. When a target signal is received, it automatically starts to verify whether the signal light timing optimization method meets the current traffic conditions, thereby improving the automation of verification, facilitating timely discovery or adjustment of the optimization direction of the signal light timing, and readjusting the signal light timing to improve traffic efficiency. In addition, the embodiment of the present invention determines the corresponding evaluation indicators based on the optimization target of the signal light timing at the last traffic intersection, making the evaluation more targeted, and multiple evaluation indicators make the evaluation more comprehensive and the evaluation results more accurate.

[0033] As an optional implementation, multiple corresponding evaluation indicators are determined based on the optimization target of the last traffic intersection signal light timing, including: performing natural language processing on the optimization target to obtain at least one keyword; when the number of keywords is less than the target number, expanding the keywords to obtain multiple related recommendation words; matching multiple first evaluation indicators corresponding to the multiple keywords and / or related recommendation words; inputting the parameters corresponding to the multiple first evaluation indicators into the digital twin model, simulating the traffic intersection situation, and adjusting the traffic intersection signal light timing to obtain the correlation between the first evaluation indicators; merging multiple first evaluation indicators whose correlation exceeds a first preset threshold to obtain multiple evaluation indicators corresponding to the traffic intersection signal light timing optimization target.

[0034] Exemplarily, in order to improve the intelligence and automation of determining the evaluation indicators in this embodiment, it is no longer dependent on the pre-set mapping relationship table between the optimization target and the evaluation indicator. It is proposed to capture the keywords of the optimization target proposed by the user through natural language processing. When the number of keywords is not enough to cover all optimization targets, the keywords are expanded through the recommendation algorithm to ensure the comprehensiveness of the evaluation indicators and further improve the accuracy of the evaluation. In addition, it should be noted that in this embodiment, the natural language processing method is used to extract the keywords in the optimization target, which can enhance the convenience and interactivity of the user, that is, it is applicable to various language habits contained in any optimization target. The keyword capture process is one of the mature natural language processing technologies, which will not be described in detail in this embodiment.

[0035] When the number of keywords is less than the target number, the keywords are expanded. The target number can be 3. The target number can be set according to the actual situation and is not limited in this embodiment. The way to expand the keywords can be to use a transportation field dictionary or an online database to find the co-occurring words of the keywords. The co-occurring words represent words that often appear with the keywords, and then identify important co-occurring words according to the network analysis tool, and use the important co-occurring words as related recommended words.

[0036] The first evaluation indicators include traffic flow, delay, queue length, traffic capacity, etc. The specific method of matching the corresponding multiple first evaluation indicators according to multiple keywords and / or related recommended words can be to establish an evaluation indicator library to classify and describe the commonly used evaluation indicators in various fields. When inputting keywords and recommended words, the most relevant evaluation indicators can be found through matching algorithms (such as vector matching, string similarity matching, etc.); it can also be a machine learning method, such as a classification algorithm, to input keywords and recommended words as features, and output the evaluation indicators learned through training data. When keywords and recommended words are used as input, the model can predict their corresponding evaluation indicators. This embodiment does not limit the specific method of matching the corresponding multiple first evaluation indicators according to multiple keywords and / or related recommended words, and those skilled in the art can determine it as needed.

[0037] The digital twin model of the traffic intersection is constructed using geometric models, semantic models and association mapping. By integrating multi-source data, comprehensive and accurate geometric information is obtained, and the construction of complex geometric models with multi-scale features and full-factor characteristics is realized. By simulating the traffic intersection conditions with the digital twin model, different signal light timing schemes can be tested virtually, saving the cost and time of actual testing. By adjusting the signal light timing of the traffic intersection in the digital twin, the correlation between the first evaluation indicators can be obtained. For example, when one of the first evaluation indicators is traffic flow and the other first evaluation indicator is queue length, when the first evaluation indicator is traffic flow, its parameter is the traffic data obtained by the sensor or monitor, and when the first evaluation indicator is queue length, its parameter is the queue length of each lane obtained by the sensor or monitor in a certain period of time. The acquired traffic flow data and queue length are input into the digital twin model. At the same time, the timing of traffic lights at traffic intersections is adjusted. Through the change of the timing of traffic lights at traffic intersections, the influence on the parameters of each first evaluation indicator is obtained, and the correlation between each first evaluation indicator is obtained. For example, when the red light time of the lane is extended by adjusting the traffic light at the traffic intersection, the queue length will increase, and its increased average length is recorded. The traffic flow will decrease, and the reduction in traffic flow is recorded. Then, according to the relationship between the average length of the queue increase and the reduction in traffic flow, the correlation between the queue length indicator and the traffic flow indicator can be obtained. The Pearson correlation coefficient can be used to obtain it, which will not be elaborated here. After determining the correlation, in order to reduce the subsequent amount of calculations, multiple first evaluation indicators whose correlation exceeds the first preset threshold are merged, that is, similar evaluation indicators are merged, to obtain multiple evaluation indicators corresponding to the signal timing optimization target of the traffic intersection. The first preset threshold can be 0.5. This embodiment does not limit the first preset threshold and can be determined according to the needs of technical personnel in this field.

[0038] As an optional implementation, after calculating the correlation of the plurality of first evaluation indicators, the method includes: marking the plurality of first evaluation indicators whose correlation is lower than the second preset threshold value to obtain the target first evaluation indicator, and evaluating the signal light timing optimization result of the traffic intersection according to the plurality of evaluation indicators and their corresponding parameters to obtain the evaluation result, including:

[0039] When there are multiple target first evaluation indicators, the strategy variable combinations of multiple target first evaluation indicators are established according to the particle swarm optimization algorithm; according to the strategy variable combination, the target space is constructed, and the target space contains all strategy variable combinations; the Pareto front is identified in the target space; the best decision result is determined in the Pareto front according to the preset constraints; the parameters corresponding to the target first evaluation indicator of the best decision result and the parameters corresponding to the target first evaluation indicator in the optimization target are quantified to obtain a differentiated result; according to the differentiated result, the final result of the optimization target of the signal light timing at the last traffic intersection in this evaluation is determined as the result of this evaluation.

[0040] Exemplarily, the correlation between multiple first evaluation indicators is lower than the second preset threshold, indicating that the multiple first evaluation indicators are relatively independent. In this embodiment, the relatively independent first evaluation indicators are referred to as target first evaluation indicators. For the target first evaluation indicator, this embodiment first obtains multiple optimization objectives, then determines multiple target first evaluation indicators, and then creates an initial particle swarm, each particle represents a set of possible solutions, that is, a combination of strategy variables, specifically a combination of traffic light timing parameters. According to the strategy variable combinations and their corresponding evaluation indicators, a target space is constructed. This space contains all possible combinations of strategy variables, and each combination is a potential solution. The PSO algorithm is used to iteratively find the optimal solution. In the PSO algorithm, the speed and position of each particle are adjusted according to the update formula, and the speed update formula is as follows:

[0041]

[0042] The position update formula is as follows:

[0043]

[0044] Among them, V id is the velocity of particle i in dimension d, X id is the position of particle i in dimension d, ω is the inertia weight, c1 and c2 are learning factors, r1 and r2 are random numbers, P id is the individual optimal position of particle i, P gd is the global optimal position, and k is the kth iteration.

[0045] The Pareto front represents a set of solutions that strike a balance between multiple objectives. These solutions are non-dominated in the objective space, that is, no other solution can outperform them on all objectives at the same time. By comparing the fitness of particles, the individual optimal solution and the global optimal solution of each particle are updated. In the process of finding the optimal solution, practical constraints can be considered, such as traffic light cycle, maximum and minimum green light time, etc. In the Pareto front, the final optimal solution is selected according to the preset constraints and the decision maker's preferences, which is the best decision result.

[0046] The parameters corresponding to the target first evaluation index of the best decision result and the parameters corresponding to the target first evaluation index in the optimization target are quantified to obtain a differentiated result. Specifically, first, the target first evaluation index parameters of the optimization strategy obtained by the last optimization target are collected through sensors or monitors. Then, the target first evaluation index parameters corresponding to the best decision result are determined. Specifically, the best decision result is input into the pre-established digital twin model to obtain the parameters corresponding to the target first evaluation index when the best decision result is implemented at the traffic intersection. After obtaining the target first evaluation index parameters corresponding to the best decision result and the target first evaluation index parameters of the optimization strategy obtained by the last optimization target, determine the impact of the difference between the best decision result and the optimization result of the last optimization target on the target first evaluation index parameters. For example, if the red light in the best decision result is 45 seconds, and the optimization strategy obtained by the last optimization target is a red light time of 30 seconds, in this case, for the two evaluation indicators of queue length and traffic flow, record their average values ​​before and after the parameter changes, and calculate the difference. The difference of the changes of multiple target first evaluation indicators caused by the difference between the best decision result and the optimization result of the last optimization target is weighted summed to obtain a quantified differentiated result. Based on the differentiation results, the final result of the optimization target of the signal timing at the last traffic intersection in this evaluation is obtained.

[0047] An embodiment of the present invention provides a signal optimization result evaluation method based on multiple indicators. The best decision result is obtained through a particle swarm optimization method. The best decision result is simulated through a digital twin to obtain a parameter corresponding to the target first evaluation indicator corresponding to the best decision result. The parameter is compared with the target first evaluation indicator parameter when the optimization strategy obtained by implementing the last optimization target is used to determine the difference. The final result of this evaluation is determined by the differentiated result. Through this method, the difference with the best decision result is used as a reference for the evaluation result, thereby improving the reliability and accuracy of the evaluation.

[0048] As an optional implementation, the parameters corresponding to the target first evaluation index of the best decision result and the parameters corresponding to the target first evaluation index in the optimization target are differentially quantified to obtain a differentiated result, including: taking the parameters corresponding to the target first evaluation index of the best decision result as the first sequence, and taking the parameters corresponding to the target first evaluation index in the optimization target as the second sequence; performing dimensionless processing on the data in the first sequence and the data in the second sequence; determining the difference value between each data in the second sequence after dimensionless processing and the data in the first sequence according to the maximum difference and the minimum difference between the data in the first sequence after dimensionless processing and the data in the second sequence; and obtaining a differentiated result according to the difference value and its corresponding weight.

[0049] Specifically, a first evaluation index of a target corresponding to the best decision result is determined, and the parameter corresponding to the first evaluation index of the target is used as the first sequence, and the parameter of the first evaluation index of the target corresponding to the optimization target is used as the second sequence, and then the first sequence data and the second sequence data are dimensionless processed to eliminate the influence of different dimensions and orders of magnitude. The dimensionless processing can include range normalization (maximum and minimum value method), averaging method, initialization method, etc.

[0050] Then, according to the maximum difference and the minimum difference between the data in the first sequence after dimensionless processing and the data in the second sequence, the difference between each data in the second sequence after dimensionless processing and each data in the first sequence is determined. The specific formula is as follows:

[0051]

[0052] Among them, χi represents the difference value of the first evaluation index of the i-th target, n represents the number of data points in the first sequence or the second sequence corresponding to the first evaluation index of the i-th target, Δ min is the minimum difference between two sequences, Δ max is the maximum difference between the two sequences, ρ is the resolution coefficient, which is 0.5, X 1i (k) is the first sequence corresponding to the first evaluation index of the i-th target, X 2i (k) is the second sequence corresponding to the first evaluation indicator of the i-th target.

[0053] When there are multiple target first evaluation indicators, a differentiated result is obtained according to the difference value of each target first evaluation indicator and the corresponding weight. The specific size of the weight can be set according to expert opinions or analysis of historical data, and this embodiment does not limit the size of the weight.

[0054] As an optional implementation, when there are multiple unlabeled first evaluation indicators, the final result of the optimization target of the traffic light timing at the previous traffic intersection in this evaluation is determined as the evaluation result of this evaluation according to the differentiation result, including:

[0055] Determine the positions of the corresponding parameters of the unmarked multiple first evaluation indicators in the evaluation indicator parameter interval preset for the optimization target; determine the evaluation values ​​of the unmarked multiple first evaluation indicators according to the positions of the corresponding parameters of the unmarked multiple first evaluation indicators in the interval; obtain the evaluation results of the unmarked multiple first evaluation indicators according to the preset first evaluation indicator weights and the corresponding evaluation values; obtain the differentiated evaluation results according to the differentiated results and their corresponding first evaluation indicator weights; obtain the final result of the optimization target of the signal timing of the previous traffic intersection in this evaluation according to the evaluation results of the unmarked multiple first evaluation indicators and the differentiated evaluation results as the evaluation result of this evaluation.

[0056] Exemplarily, the parameters corresponding to the unlabeled multiple first evaluation indicators are compared with the preset evaluation indicator parameter interval to determine the position of each parameter in the interval. According to the position of the evaluation indicator parameter in the preset interval, an evaluation value is determined for each unlabeled evaluation indicator. This evaluation value can be a direct mapping based on the parameter position, or it can be a result obtained by a pre-trained scoring model. According to the preset first evaluation indicator weight and the corresponding evaluation value, the evaluation result of each unlabeled evaluation indicator is calculated using a weighted summation method. According to the differentiated result and its corresponding first evaluation indicator weight, the differentiated evaluation result is calculated, for example, the differentiated result of each evaluation indicator is weighted summed. The evaluation results of the unlabeled multiple first evaluation indicators are combined with the differentiated evaluation results to obtain the final result of the optimization target of the last traffic intersection signal light timing in this evaluation. The final evaluation result is used as the output of this evaluation, and this result can be used to judge the applicability and effectiveness of the last optimization target in the current situation.

[0057] The embodiment of the present invention provides a signal optimization result evaluation method based on multiple indicators, which combines the evaluation values ​​of multiple unlabeled first evaluation indicators and the evaluation values ​​of multiple labeled first evaluation indicators. Its evaluation dimension is more complete and comprehensive, and the evaluation result is more accurate.

[0058] As an optional implementation, a signal optimization result evaluation method based on multiple indicators further includes:

[0059] Obtain multiple evaluation results; perform trend analysis on the multiple evaluation results to obtain a trend analysis graph, which includes points corresponding to the multiple evaluation results and the overall trend; and determine the applicability of the optimization target of the previous traffic intersection signal timing under the current traffic intersection conditions based on the trend analysis graph.

[0060] Exemplarily, statistical software or programming languages ​​are used to perform trend analysis on the data. Use data analysis tools to draw a trend analysis graph, which should contain: data points, with the results of each evaluation as points on the graph; trend lines, which show the overall trend of the data, which can be linear or nonlinear. Observe the trend analysis graph, analyze the distribution of data points and the direction of trend lines, and determine the changing trend of the evaluation results over time. Evaluate whether the trend is significant, whether it continues to rise or fall, or whether there are fluctuations. Based on the results of the trend analysis graph, evaluate the applicability of the optimization goal of the last traffic intersection signal timing to the current traffic intersection situation: if the trend shows continued improvement, it may indicate that the current optimization goal is still applicable. If the trend shows that the performance indicator has not improved significantly or has declined, the optimization goal may need to be adjusted. If the trend is unstable or fluctuates greatly, further analysis of the data may be required to determine the effectiveness of the optimization goal.

[0061] As an optional implementation, the trend analysis graph includes points corresponding to multiple evaluation results and overall trend data, and determining the applicability of the optimization target of the signal timing of the previous traffic intersection under the current traffic intersection based on the trend analysis results, including: obtaining points corresponding to multiple evaluation results of the trend analysis graph, and determining the slope data of the line between the point in the trend analysis graph and the adjacent points; determining a first suitability parameter based on the slope data of the line between the point in the trend analysis graph and the adjacent points; determining a second suitability parameter based on the overall trend data; and determining the applicability of the optimization target of the signal timing of the previous traffic intersection under the current traffic intersection based on the first suitability parameter and the second suitability parameter.

[0062] Exemplarily, data points corresponding to each evaluation cycle are extracted from the trend analysis graph, and these points represent the evaluation results at different time points. For each data point in the trend analysis graph, the slope of the line between it and the adjacent point is calculated. According to the calculated slope data, the first fitness parameter is determined. This parameter can be: average slope. The overall trend line is used to determine the second fitness parameter, and the second fitness parameter can be the slope of the overall trend line. Combined with the first fitness parameter and the second fitness parameter, the fitness of the optimization target of the last traffic intersection signal timing in the current situation is comprehensively evaluated. Specifically, the difference between the average slope and the slope of the overall trend line is calculated. When the difference between the average slope and the slope of the overall trend line is within a preset range, the corresponding fitness is searched in the mapping table of the slope of the overall trend line and the fitness. The larger the slope in the mapping table, the higher the corresponding fitness. The specific value of the mapping relationship can be determined according to the actual situation; when the difference between the average slope and the slope of the overall trend line exceeds the preset range, its fitness can be a fixed value, which is used to indicate that the trend is unstable and to remind the user that the optimization target needs to be adjusted.

[0063] The embodiment of the present invention provides a signal optimization result evaluation method based on multiple indicators. The method determines the suitability by analyzing the trends of multiple evaluation results, thereby avoiding the one-sidedness caused by determining the suitability by a single evaluation, and avoiding the problem that the last optimization goal can be met most of the time, but the time of initiating the evaluation is a certain period of time, resulting in the evaluation result not meeting the last optimization goal, thereby improving the reliability of the suitability results.

[0064] This embodiment provides a signal optimization result evaluation device based on multiple indicators, including:

[0065] A target acquisition module, for obtaining the optimization target of the last traffic intersection signal timing from the historical records in response to the target signal;

[0066] An index determination module, used to determine a plurality of corresponding evaluation indicators according to the optimization target of the last traffic intersection signal light timing;

[0067] A parameter acquisition module, used for acquiring parameters corresponding to the evaluation indicators based on the plurality of evaluation indicators, wherein the parameters are obtained based on vehicle flow and / or pedestrian flow data at the traffic intersection;

[0068] The result determination module is used to evaluate the signal light timing optimization result of the traffic intersection according to the plurality of evaluation indicators and their corresponding parameters to obtain the evaluation result.

[0069] The present application also provides an electronic device, such as Figure 2 As shown, a processor 501 and a memory 502, wherein the processor 501 and the memory 502 may be connected via a bus or other means.

[0070] The processor 501 may be a central processing unit (CPU). The processor 501 may also be other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, or a combination of the above chips.

[0071] The memory 502 is a non-transient computer-readable storage medium that can be used to store non-transient software programs, non-transient computer executable programs and modules, such as program instructions / modules corresponding to the signal optimization result evaluation method based on multiple indicators in the embodiment of the present invention. The processor executes various functional applications and data processing of the processor by running the non-transient software programs, instructions and modules stored in the memory.

[0072] The memory 502 may include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application required for at least one function; the data storage area may store data created by the processor, etc. In addition, the memory may include a high-speed random access memory, and may also include a non-transient memory, such as at least one disk storage device, a flash memory device, or other non-transient solid-state storage device. In some embodiments, the memory 502 may optionally include a memory remotely arranged relative to the processor, and these remote memories may be connected to the processor via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0073] The one or more modules are stored in the memory 502, and when executed by the processor 501, the following is performed: Figure 1 The signal optimization result evaluation method based on multiple indicators in the illustrated embodiment.

[0074] For details of the above electronic equipment, please refer to Figure 1 The corresponding related descriptions and effects in the illustrated embodiments can be understood and will not be repeated here.

[0075] This embodiment also provides a computer storage medium, which stores computer executable instructions, and the computer executable instructions can execute the signal optimization result evaluation method based on multiple indicators in any of the above method embodiments. Among them, the storage medium can be a disk, an optical disk, a read-only memory (ROM), a random access memory (RAM), a flash memory (Flash Memory), a hard disk (HDD) or a solid-state drive (SSD), etc.; the storage medium can also include a combination of the above types of memory.

[0076] Finally, it should be noted that the above preferred embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail through the above preferred embodiments, those skilled in the art should understand that various changes can be made in form and details without departing from the scope defined by the claims of the present invention.

Claims

1. A signal optimization result evaluation method based on multiple indicators, characterized in that: include: In response to the target signal, obtaining the optimization target of the last traffic intersection signal timing from the historical records; Determine a plurality of corresponding evaluation indicators according to the optimization target of the last traffic intersection signal light timing; Based on the plurality of evaluation indicators, obtaining parameters corresponding to the evaluation indicators, wherein the parameters are obtained based on vehicle flow and / or pedestrian flow data at the traffic intersection; According to the plurality of evaluation indicators and their corresponding parameters, the signal light timing optimization result of the traffic intersection is evaluated to obtain the current evaluation result.

2. The signal optimization result evaluation method based on multiple indicators according to claim 1 is characterized in that: The method of determining the corresponding multiple evaluation indicators according to the optimization target of the last traffic intersection signal light timing includes: Perform natural language processing on the optimization target to obtain at least one keyword; When the number of keywords is less than the target number, the keywords are expanded to obtain multiple related recommended words; Matching corresponding multiple first evaluation indicators according to multiple keywords and / or related recommendation words; Inputting the parameters corresponding to the plurality of the first evaluation indicators into the digital twin model to simulate the traffic intersection conditions; Adjust the timing of traffic lights at traffic intersections to obtain the correlation between the first evaluation indicators; Multiple first evaluation indicators whose correlation exceeds a first preset threshold are combined to obtain multiple evaluation indicators corresponding to the signal light timing optimization target of the traffic intersection.

3. The signal optimization result evaluation method based on multiple indicators according to claim 2 is characterized in that: After calculating the correlation of the plurality of first evaluation indicators, the method includes: marking the plurality of first evaluation indicators whose correlation is lower than the second preset threshold value to obtain the target first evaluation indicator; and evaluating the signal light timing optimization result of the traffic intersection according to the plurality of evaluation indicators and their corresponding parameters to obtain the evaluation result, including: When there are multiple target first evaluation indicators, a strategy variable combination of multiple target first evaluation indicators is established according to the particle swarm optimization algorithm; According to the strategy variable combination, a target space is constructed, where the target space includes all strategy variable combinations; identifying a Pareto front in the target space; Determine the best decision result in the Pareto front according to the preset constraints; Quantify the difference between the parameter corresponding to the target first evaluation index of the optimal decision result and the parameter corresponding to the target first evaluation index in the optimization target to obtain a differentiated result; According to the differentiation results, the final result of the optimization target of the signal light timing at the last traffic intersection in this evaluation is determined as the result of this evaluation.

4. The signal optimization result evaluation method based on multiple indicators according to claim 3 is characterized in that: The step of quantifying the difference between the parameter corresponding to the target first evaluation index of the optimal decision result and the parameter corresponding to the target first evaluation index in the optimization target to obtain a differentiated result includes: The parameters corresponding to the target first evaluation index of the optimal decision result are used as the first sequence, and the parameters corresponding to the target first evaluation index in the optimization target are used as the second sequence; Performing dimensionless processing on the data in the first sequence and the data in the second sequence; Determine the difference between each data in the second sequence after the dimensionless processing and the data in the first sequence according to the maximum difference and the minimum difference between the data in the first sequence after the dimensionless processing and the data in the second sequence; A differentiation result is obtained according to the difference value and its corresponding weight.

5. The signal optimization result evaluation method based on multiple indicators according to claim 3 is characterized in that: When there are still multiple unlabeled first evaluation indicators, determining the final result of the optimization target of the traffic light timing at the last traffic intersection in this evaluation as the evaluation result of this evaluation according to the differentiation result includes: Determine the positions of the unmarked first evaluation index corresponding parameters in the evaluation index parameter interval preset in the optimization target; Determining evaluation values ​​of the unlabeled plurality of first evaluation indicators according to positions of the parameters corresponding to the unlabeled plurality of first evaluation indicators in the interval; Obtaining evaluation results of multiple unlabeled first evaluation indicators according to preset first evaluation indicator weights and corresponding evaluation values; Obtaining a differentiated evaluation result according to the differentiated result and its corresponding first evaluation index weight; According to the evaluation results of the unlabeled multiple first evaluation indicators and the differentiated evaluation result, the final result of the optimization target of the traffic light timing at the previous traffic intersection in this evaluation is obtained as the evaluation result of this time.

6. The signal optimization result evaluation method based on multiple indicators according to claim 1 is characterized in that: Also includes: Obtain multiple evaluation results; Performing trend analysis on the evaluation results for multiple times to obtain a trend analysis graph, wherein the trend analysis graph includes points corresponding to the evaluation results for multiple times and an overall trend; According to the trend analysis diagram, the applicability of the optimization target of the traffic light timing at the last traffic intersection to the current traffic intersection situation is determined.

7. The signal optimization result evaluation method based on multiple indicators according to claim 6 is characterized in that: The trend analysis graph includes points corresponding to multiple evaluation results and overall trend data. The determination of the applicability of the optimization target of the last traffic intersection signal timing under the current traffic intersection situation based on the trend analysis result includes: Obtain points corresponding to multiple evaluation results of the trend analysis graph, and determine slope data of the lines connecting the points in the trend analysis graph and adjacent points; Determine a first fitness parameter according to slope data of a line connecting a point in a trend analysis graph and adjacent points; determining a second fitness parameter based on the overall trend data; Based on the first suitability parameter and the second suitability parameter, the suitability of the optimization target of the traffic light timing at the previous traffic intersection under the current traffic intersection conditions is determined.

8. A signal optimization result evaluation device based on multiple indicators, characterized in that: include: A target acquisition module, for obtaining the optimization target of the last traffic intersection signal timing from the historical records in response to the target signal; An index determination module, used to determine a plurality of corresponding evaluation indicators according to the optimization target of the last traffic intersection signal light timing; A parameter acquisition module, used for acquiring parameters corresponding to the evaluation indicators based on the plurality of evaluation indicators, wherein the parameters are obtained based on vehicle flow and / or pedestrian flow data at the traffic intersection; The result determination module is used to evaluate the signal light timing optimization result of the traffic intersection according to the plurality of evaluation indicators and their corresponding parameters to obtain the evaluation result.

9. An electronic device, comprising: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the steps of the signal optimization result evaluation method based on multiple indicators as described in any one of claims 1 to 7.

10. A computer storage medium having computer instructions stored thereon, characterized in that: When the instruction is executed by the processor, the steps of the signal optimization result evaluation method based on multiple indicators described in any one of claims 1-7 are implemented.

Images